Implementation of Blind and Non-blind Deconvolution for Restoration of Defocused Image

Now a days the use of digital camera in day-today life is increased and along with which new challenges are rising demanding the quality improvement of captured images. Further, these quality improvement challenges and problems of captured images needs to be taken into account and required to be resolved continuously. One of the challenging tasks in the field of image processing comes out as restoration of defocused images by removing blur or noise. The blur in the defocused image generates because of different camera parameters, lens settings and observation errors while capturing the image. This paper addresses the challenge of removal of blur from the captured image using deconvolution techniques for restoration of defocused image. Use of blind and non-blind deconvolution techniques are used in this work for restoration of defocused image. The major problem is no prior knowledge of point spread function in blind deconvolution compared to non-blind deconvolution. It is also addressed in this work which is useful for further usage in computer vision and artificial intelligence. Finally, the results after implementation are presented in terms of blur type, performance parameters like signal to noise ratio (SNR), mean square error (MSE) and peak signal to noise ratio (PSNR). On the basis of these performance parameters, blind and non-blind deconvolution methods are also compared.